alt Hacker NewsCould be amazing, but it's hard to judge if it will really work with say a 27 B model or larger. We can already get pretty good speed with a 2B model.
Could be amazing, but it's hard to judge if it will really work with say a 27 B model or larger. We can already get pretty good speed with a 2B model.
thanks! we explain how it scales to larger models in the last section the OP blog post